Close

Impact of Size

13 May Posted by in Market Microstructure | Comments Off on Impact of Size
Impact of Size
 

 

Even if  your not an institution, there may come a point where the size of your trades impacts the price you receive.  What size threshold behooves you to recognize (ala Heisenberg), that your interaction (interference) with the market, changes the market? In order to cleverly adjust order entry for optimal fills, market microstructure assessment is essential. Measuring the magnitude of market impact of certain size orders, is prerequisite to any corresponding adjustment to an order placement distribution, as well as using alternative liquidity sources, such as dark liquidity pools. 

In general, market depth (where volume is plotted as a function of price) has something like a “V” shaped distribution. In other words, the volume distribution of bids and offers, say in an order book, is “somewhat” symmetric around the current price, where the volume of bids and offers approaches a minimum. But market depth (as typically depicted) is not a complete picture, as it only reflects the limit orders that are resting. Order placement in a sequestered fashion may in fact draw “new” buyers and sellers into the market.

 One way to  measure the impact, that a certain order size, has on a market, is to observe the change, in the log midprice of the underlying instrument,  due to a trade of a certain size or volume, say n. It is currently accepted that market impact, scales according to the square root of the trade size. Further distribution analysis suggests, that a market impact parameter, say “alpha”,  more precisely modifies the square root scaling of impact. What is “alpha”? Alpha is equal to volatility, or sigma, divided by the square root of  the number of units (eg. shares, contracts, lots, etc.) traded per a unit of trading time,say “u” (u=#of units/time=trade density).Therefore, alpha =sigma/sq. rt of u.  Measuring the degree of impact a particular trade size has on a market, is a first step to facilitate those order placement distributions, which are most efficient, thereby reducing transactional fees, and allowing for optimal return capture.